[Arm-netbook] apologies

[luke.leighton]

On Tue, Nov 12, 2013 at 1:58 PM, joem <joem at martindale-electric.co.uk> wrote:
> On Tue, 2013-11-12 at 12:09 +0000, luke.leighton wrote:
>> On Mon, Nov 11, 2013 at 2:10 AM, Christopher Havel
>> <laserhawk64 at gmail.com> wrote:
>> > On 11/10/2013 9:08 PM, luke.leighton wrote:
>> >>
>> >> apologies to everyone on the list - you may have noticed i'm using
>> >> brief and curt language at the moment.  it's not personal: i just have
>> >> far less time available and have to be very focussed.
>> >> l.
>> >>
>> >> _______________________________________________
>> >> arm-netbook mailing list arm-netbook at lists.phcomp.co.uk
>> >> http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook
>> >> Send large attachments to arm-netbook at files.phcomp.co.uk
>> >
>> > Sorry to hear you're so busy. Can I help? :D
>>
>>  yes.  i thought of something.
>> http://permalink.gmane.org/gmane.comp.hardware.netbook.arm/7597
>>
>>  could you investigate, find some "educational" style electronics
>> explanations online, as to why henrik's recommendation to use a
>> pull-up resistor to 3.3V and a reversed-diode between the Tx UART and
>> incoming Rx actually works?  i'm attaching a small diagram showing the
>> circuit.
>>
>>  it can then be added here:
>> http://elinux.org/Embedded_Open_Modular_Architecture/EOMA-68#Requirements_for_UART
>>
>>  also i think.... to avoid too big a voltage-drop it should be a zener
>> diode otherwise it might interfere with the thresholds.
>
> Henrik's idea should work - the diode should be a schottky diode with
> 0.2V to 0.3V drop as hinted in the drawing otherwise a 0.7V drop with a
> normal diode may be too high to allow a logic 0 to be registered.
>
> The down side of the circuit is that the pull up resistor needs to be
> connected to the 3.3V line, which is inside the EOMA and not accessible.

 ... but if the circuit you did can do it with external resistors then
that can also be arranged.  as long as the pull-up voltage is greater
than a 3.3v logic-level that should be enough.  that leaves... 1.3v of
wiggle-room... so splitting the resistors into say 1k and 10k in
series and bringing out the pull-up in between the two of them should
be sufficient, right?

> Also - if higher baud rates don't work try decreasing the value of the
> resistor.

 ok.  thanks joe.

l.